JP4898342B2 - Color conversion method and color conversion system - Google Patents

Description

  This exemplary embodiment relates to the field of digital imaging. This finds particular application associated with correcting colorant input values due to the effects of other colorants present in image rendering devices such as printers. However, the methods and systems disclosed herein are applicable to other rendering techniques, such as color image display.

  Image non-uniformity occurs in the output image of digital imaging devices such as copiers, scanners, and printers for a variety of reasons. In a gray uniform patch, streaks, stripes and bands can appear as variations in gray levels. In general, “gray” refers to the optical density or area coverage value of any single color separation layer, whether the toner is black, cyan, magenta, yellow, or some other color.

US Pat. No. 6,760,056

  One-dimensional tone reproduction curves (TRC) are widely used in digital imaging as a means to compensate for non-linearities introduced by individual imaging devices. Some success has been achieved in correcting the spatial uniformity of a monochrome image with respect to the formation of stripes and streaks and stripes by appropriate modification to the tone reproduction curve. In color imaging, correction of spatial non-uniformities such as streaks has proven to be more difficult due to the interaction between the colorants.

  In one aspect, a method for converting colors includes: a) establishing a plurality of tone reproduction curves for a first color separation of a plurality of color separations in a digital image, each of the color reproduction curves. A colorant interaction between a first colorant that renders the first color separation and at least one second colorant that renders at least a second color separation of the plurality of color separations. Each of the tonal reproduction curves includes a modified input value corresponding to an input value representing color separation at least for a fixed input value of the second color separation; and b) a predetermined representing the first color separation. For digital image pixels having an input value and at least a predetermined input value representing a second color separation, a fixed value representing the second color separation from a plurality of tone reproduction curves. Selecting at least one tone reproduction curve whose input value is close to a predetermined input value representing the second color separation; and c) representing the first color separation for each of the selected tone reproduction curves. Determining a modified input value representing a first color separation corresponding to the input value; and d) if two or more tone reproduction curves are selected, the first calculated from the selected reproduction curve Determining a combined corrected input value representative of the color separation of the pixel as a function of the corrected input value representative of the color separation.

  Color conversion can be performed for each of a plurality of color separations in the digital image. Establishing a plurality of tone reproduction curves may include establishing a set of tone reproduction curves for each of the plurality of regions of the image, one of the regions including pixels, the selected tone reproduction. A curve can be identified with respect to at least the region containing the pixel. The selected tone reproduction curve may also be identified with respect to at least one region adjacent to the region containing the pixel, and the method may be divided from the selected tone reproduction curve with respect to that region and with respect to at least one adjacent region. The method may further include determining a modified input value that represents the color separation of the pixel as a function of the modified color separation modified input value. The selected tone reproduction curve can be identified with respect to a plurality of regions adjacent to the region containing the pixels. The first color separation combined corrected input value for the pixel as a function of the first color separation corrected input value determined from the tone reproduction curve selected for that region and for at least one adjacent region. Determining the first color separation modified input value determined from the selected tone reproduction curve for that region and for at least one adjacent region that is a function of pixel location. And determining an average of the weighted color separation modified input values determined from the selected tone reproduction curve.

  The plurality of regions can include at least four regions. The plurality of regions can include regions that are equally spaced in at least one of a processing direction and a direction orthogonal to the processing.

  There can be at least three color separations. Identifying the plurality of selected tone reproduction curves can include identifying four tone reproduction curves.

  Determining the combined modified input value representing the first color separation of the pixel includes determining a weighted average of the modified input values representing the first color separation determined from the selected tone reproduction curve. be able to.

  Determining a combined modified input value representing a first color separation of a pixel as a function of a modified input value representing a first color separation determined from a selected tone reproduction curve is within a digital image. Determining a combined modified input value representative of the first color separation of the pixel as a function of the pixel position.

  Establishing a plurality of tone reproduction curves can include establishing a set of tone reproduction curves for each of a plurality of regions in an image. Identifying the selected color reproduction curve may include identifying the selected color reproduction curve from a set of color reproduction curves relating to the region where the pixel is located among the plurality of regions.

  Each tone reproduction curve can be represented by a plurality of basis vectors.

  The fixed input value of the at least one second color separation of the selected tone reproduction curve can define a range of predetermined input values for the second color separation for the pixel of interest.

  In another aspect, a system for converting colors includes a memory that stores a plurality of tone reproduction curves for each of a plurality of color separations. Each of the tone reproduction curves includes a colorant interaction between a first colorant that renders a color separation and at least one second colorant that renders at least one second color separation of the plurality of color separations. Represents the action. Each of the tone reproduction curves includes a modified input value corresponding to an input value representing color separation at least in the fixed input value of the second color separation. The system further includes a processing component that accesses the memory for the pixel of the digital image, wherein the fixed input value for at least the second color separation is the second color separation input value for that pixel. Identifying a subset of the tone reproduction curve from a plurality of tone reproduction curves defining a range of colors, and determining, from each subset of the tone reproduction curve, a modified input value representing a color separation corresponding to an input value representing the color separation is selected and selected. Determining a combined corrected input value representing the color separation of the pixel as a function of the corrected input value representing the color separation determined from the tonal reproduction curve.

  An image rendering device, such as a Xerographic System, may include a plurality of colorants for rendering an image and a system for converting colors as described above.

  In another aspect, a method of imaging a color includes: a) converting a first color separation input value in a digital image for each of a plurality of pixels of interest in the digital image to Representing the interaction between a first colorant associated with one color separation and at least one second colorant associated with at least one second color separation, the transformation Generating a modified input value representative of the first color separation; and b) converting the input value of the second color separation in the digital image to between the second colorant and at least the first colorant. Representing the interaction at, wherein the transformation produces a modified input value representing the second color separation. The method of imaging a color further includes forming an image using at least the first and second colorants utilizing modified input values representing the first color separation and the second color separation.

  An aspect of an exemplary embodiment is for converting an input gray level (“input value”) into a modified input gray level (“modified input value”) that can be used by an image rendering device in rendering the pixels of the image. Relates to a method and system. The modified input value represents the colorant interaction between one colorant used in rendering the image and at least one other colorant of the plurality of colorants. The input value may be the target output value of the image when rendered by the image rendering device, or may have undergone processing prior to the conversion step.

  In general, "Colorant" is used to render a specific color separation combined with one or more other colorants to achieve the color of the image throughout the spectrum when forming a multicolor image. Refers to media used in Thus, each color separation can have its own corresponding colorant. In the case of toner, the colorant is usually cyan, magenta, yellow, and sometimes black (CMYK) toner. In an ink jet printer, the colorant is ink. In the case of a display, the colorant can be red, green, and blue light. Typically, input values and modified input values are represented numerically, the higher the numerical value, the greater the saturation or gray level of its color separation in the resulting image. The modified input value can be used as an input value for the imaging device or can be further processed prior to such use. Input values for each of the color separations can reach the system in the form of a bitmap. The image rendering device uses the modified input values (directly or after further processing) to determine the image pixel, such as the total coverage of the colorant (in the case of toner) and the intensity of the colorant (in the case of display). Determine the gray level of the colorant used in forming.

  A tone reproduction curve (TRC) is generally used to correct for device nonlinearities in image rendering devices such as marking engines that produce output levels that are not linearly proportional to a specified input gray level. A TRC generally includes a series of cells that provide a modified input gray level for each set of input values corresponding to a desired output gray level of the imaging device. Some cells in the TRC can contain measurements, while others contain values appropriately interpolated between measurements. The engine response curve describes the relationship of the input gray level to the printed gray level or the displayed gray level. The TRC that corrects the gray level response of the engine has an inversely proportional relationship with the engine response curve. The term TRC is intended to include data stored in the form of engine response curves.

  In one aspect, a method for correcting colors includes establishing one or more tone reproduction curves (TRC) for each of a plurality of color separations for rendering an image, each of the TRCs comprising: A colorant associated with the color separation of interest (referred to herein as the first color separation or PCS (Primary Color Separation)) and at least one other color separation of the plurality of color separations (herein Represents a non-linear colorant interaction with (one or more) second color separations or SCS (Secondary Color Separation). Each of the TRCs has an input value representing a PCS and a corresponding modification in one or more fixed input values representing the SCS (s) associated with the one or more second colorants present. Contains input values. These TRCs can be one-dimensional with a set of TRCs specific to color separation (hereinafter referred to as TRC with color index or CITRC (Color-Indexed TRC)), respectively. Color indexed TRCs provide input values and modified input values at separate fixed input values of the SCS (s). A selected CITRC of the plurality of CITRCs is accessed to determine a modified input value representing a color separation corresponding to a desired target level in the presence of a particular amount of second colorant (s). The

  If multiple CITRCs are used, the output can be a function such as a weighted average of values obtained from more than one CITRC. The modified input value can be unique to one region of the image to provide color correction specific to the spatially dependent colorant. Depending on the location of the pixel being corrected, the corrected input value can be a function of one or more of the corrected input values specific to the region.

  In some aspects of the exemplary embodiments, the system and method provides colorant over a desired uniformity space, such as correcting pixels of the image with respect to device nonlinearities in the processing direction or in a direction orthogonal to the processing. Makes it possible to correct the non-uniformity of the. A particular aspect relates to correction of color image streaks and fringes in printed matter. This specification specifically refers to the desired uniform space as being a spatial dimension, such as X or Y (eg, the direction orthogonal to the processing of the image and the processing direction) on the image, for example. However, the desired uniformity space may include a dimension in time, a dimension from device to device, or other variables. For example, the time dimension can be used to correct for variability that can be predicted with reasonable accuracy, such as known variability or variability that occurs during a day for a given marking engine in a printing system. . The device-to-device dimension can incorporate CITRC for one or more marking engines.

  The CITRC can be spatially dependent so that the correction applied to each pixel depends not only on the input value for that pixel, but also on the row or column address of that pixel. The correction can be applied equally to all rows to correct column-to-column variations, or it can be applied equally to all columns to correct row-to-row variations. It can also be applied to both rows and columns to correct both types of variations. In the TRC, as shown in FIG. 1, Y1, Y2, Y3, etc. correspond to spatial positions in the Y direction (processing direction), and X1, X2, X3, etc., correspond to the X direction (direction orthogonal to the processing). ) In the X and Y dimensions corresponding to the spatial position in FIG.

FIG. 1 represents a two-dimensional image 10. The image 10 can be a digital image rendered on a print medium, such as paper, or a digital image displayed on a screen or otherwise, and includes a plurality of pixels 12. Each of the plurality of regions 14 in the image 10 can be identified by X and / or Y coordinates, such as (X1, Y1). In FIG. 1, region 14 is shown as a spatially distinct, adjacent region of equal size and including a plurality of pixels in both X and Y directions. However, the regions 14 can be of different sizes and spaced from each other. For example, there can be N regions in the X direction and M regions in the Y direction. N can be from 1 to N _max , where N _max is the total number of pixels in the X direction and M can be from 1 to M _max , where M _max is , The total number of pixels in the Y direction. In general, N and M can be from about 1 to 50, where at least one of N and M is greater than 1, for example, at least one of N and M is from about 2 to about 20 Yes, for example at least 8. In one embodiment suitable for streak and fringe correction, N can be about 16, with M = 1 corresponding to a region about 1 to 3 cm wide. The number of regions 14 can depend on the memory available to store the corresponding TRC and the processing speed of the processing equipment used with the TRC. Furthermore, the width of the region can depend on the degree of non-uniformity change.

  The method disclosed herein can be used to input color separations due to the effects of other color separations on one region equal to the entire image, ie other color separations when both N and M are equal to 1. It is also applicable to correcting values. In such a case, the existing non-uniformity is not considered. Thus, although this specification specifically refers to region 14 as being smaller than one area of image 10, this region may be the entire image.

  TRC with color index

  In one aspect of the exemplary embodiment, each of the regions 14 is associated with a plurality of corresponding color indexed TRCs 16. Specifically, each of the PCSs is associated with a set of CITRCs 16 and each CITRC corresponds to a fixed (one or more) amount of other ("second") color separations present. To do. For example, in the case of cyan (C), magenta (M), and yellow (Y) as PCS, there may be, for example, at least 9 or at least 12 CITRCs 16 for each PCS, and one embodiment There can be more than 16 CITRCs for each PCS, each of which corresponds to a separate set of fixed input values for the other two color separations. FIG. 1 shows two of CITRCs 16 representing cyan as an example.

The number N _c ⁱ of color indexed TRCs for the i-th PCS can be expressed as:

(Equation 1)
i = 1, 2,. . . , Z + 1, N _c ⁱ = [N _S1 ] ⁱ × [N _S2 ] ⁱ . . . × [N _Sz ] ⁱ

In equation (1), z is the total number of SCSs considered, and for the i th first color to be corrected, N _S1 ⁱ is the fixed level of the first SCS of the SCSs. N _S2 ⁱ is the number of fixed levels of the second SCS of the SCS, etc., and N _S1 ⁱ , N _S2 ^i, etc. are independently about 2 to about 20 Can be up to. The total number T of color indexed TRCs for each area can be obtained by the equation (2).

(Equation 2)
T = N _c ¹ + N _c ² +. . . + N _c ^{(z + 1)}

Thus, for example, in a system that employs three color separations (eg, cyan (C), magenta (M), and yellow (Y)) and four fixed levels of each SCS, each region For 14, Ns _j ⁱ = 4 and N _c ⁱ = 16 (for i = 1, 2, 3, and j = 1, 2). Therefore, overall, T = [4 × 4] + [4 × 4] + [4 × 4] = 48 TRCs with color indexes. The number of CITRCs representing each color separation can depend in part on the available memory storage and on the extent to which the second colorant affects their CITRC. When black (K) is one of the first colorants, it is unnecessary for some marking engines to provide a TRC with a color index that corrects for different amounts of black colorant. In general, a color indexed TRC can be provided for all or fewer than all of the color separations and associated colorants used in the marking engine. Therefore, it is conceivable that a TRC with a color index that takes black into account can be provided instead, but in the case of the CMYK system, cyan (C), magenta (M), and yellow (Y) as the second colorant. Only a TRC with a color index that corrects the interference may be used.

  A fixed input value representing each SCS includes a first value at or near the upper limit of the color separation range, a second value at or near the lower limit of the range, and the first value And at least one value intermediate between the second value and the second value. When color separation is represented on an 8-bit color scale from 0 to 255, for example, as shown in Table 1 below, the input values magenta (M) and yellow (Y There can be 16 CITRCs for the first color cyan (C) with fixed).

(Table 1)
Table 1: Exemplary CITRC for cyan (C)
1. M = 0, Y = 0 5. M = 0, Y = 160 9. M = 0, Y = 224 13. M = 0, Y = 255
2. M = 160, Y = 0 6. M = 160, Y = 160 10. M = 160, Y = 224 14. M = 160, Y = 255
3.M = 224, Y = 0 7.M = 224, Y = 160 11.M = 224, Y = 224 15.M = 224, Y = 255
4. M = 255, Y = 0 8. M = 255, Y = 160 12. M = 255, Y = 224 16. M = 255, Y = 255

  In this embodiment, the fixed input level of the SCS includes these color separation minimum and maximum values (0 and 255) and two intermediate values, although other fixed values can also be selected. I want you to understand. Since the impact of SCS tends to become more pronounced at higher gray levels of SCS, more of the fixed level is selected to fit within the upper half of the range than the lower half of the range of color values can do.

  Similar CITRCs are provided for each of the other PCS, magenta (M), and yellow (Y). In the case of magenta (M), the input values of cyan (C) and yellow (Y) for each CITRC are fixed, and in the case of yellow (Y), the input values of cyan (C) and magenta (M) Is fixed. Each of the 16 CITRCs for a given PCS includes a plurality of input values representing its color separation and a corresponding modified input value. The values in CITRC can be relatively sparse. For example, as shown for PCS cyan (C), magenta (M), and yellow (Y) in FIG. 2, in each of the 16 CITRCs representing each color separation, at least 4 and one There may be at least 6 input values in an embodiment, or even about 9 input values and a corresponding modified input value in a particular embodiment. These input values are used as addresses or indexes to access the modified input values that are in the lookup table or in functional form.

  Although FIG. 1 graphically illustrates two exemplary CITRCs 16 for cyan (C) with respect to region 14, in practice CITRC may be used as input values and modified input values, or as described in more detail below. Can be stored as part of a function that includes weights for the CITRC. Each CITRC can be described as “one-dimensional” in the sense that it includes coordinates in a single plane. Thus, for each input value, there is a single corresponding modified input value. When a desired input value is input into CITRC, a corresponding modified input value (representing cyan (C) in the illustrated case) is generated.

  There are more possible SCS input values (256 in the illustrated embodiment) than there are CITRCs present. As will be described in more detail later, the intermediate value can be obtained by interpolation.

In one embodiment, the modified input values representing cyan, magenta, and yellow for a given pixel are determined from CITRC 16 for the region 14 where that pixel is located. In another embodiment, these modified input values are interpolated from the region 14 where the pixel is located and one or more of the closest regions, eg, CITRC of up to four regions in total. Determined by. For example, as shown in FIG. 1, the cyan corrected input values of the pixel 12 located in the lower right quadrant of the region (X4, Y2) are the regions (X4, Y2), (X4, Y3). , (X5, Y2), and (X5, Y3) can be the weighted average of the values of C _mod determined from the CITRC. The weighting appropriate for different C _mod values can be inversely proportional to the pixel distance from the midpoint of each region. If the entire image 10 is a single region 14, the value of C _mod is not a function of the value of adjacent C _mod which is indexed with respect to regions, therefore at this stage interpolation is not required at all.

  FIG. 3 illustrates an exemplary image rendering device in the form of a digital printing system 20, which includes at least one color correction system (such as a color indexed TRC 16 of the type described herein). CCS) 22. Image data 24 representing the image to be printed is received by an image processing system (IPS) 28, which can incorporate what is known in the art as DFE (digital front end). Image processing system 28 processes received image data 24 to generate image data 30 in a format that can be processed by one or more output devices 32, 34 (eg, (one or more) marking engines). Image data 30 is received by an appropriate marking engine 32, 34 for printing the image. The image processing system 28 can receive the image data 24 from an input terminal 36 such as a scanner that captures an image from the original document, a computer, a network, or any similar or equivalent image input terminal that communicates with the image processing system 28. . In some aspects, some or all of the functionality of IPS is provided by a marking engine, eg, CCS 22.

  The marking engines 32, 34 shown are electrophotographic engines, but the CCS 22 can be used with a variety of digital forms of copying and printing equipment. The output devices 32, 34 can be color xerographic printers, ink jet printers, and the like.

  Each of the marking engines 32, 34 can incorporate various xerographic subsystems 40, 42 to form an image, transfer the image to a print medium such as a piece of paper, and fuse the image. . In the case of xerographic devices, the marking engine typically includes a charge retaining surface, such as a rotating photoreceptor. An image is formed on the surface of the photoreceptor. Arranged around the photoreceptor is a xerographic subsystem, which is a charging station and an exposure associated with each charging station that forms a latent image on the photoreceptor. And a developing unit associated with each charging station for developing a latent image formed on the surface of the photoreceptor by applying a colorant such as toner to obtain a toner image; A transfer unit that transfers the toner image formed in this way to the surface of the printing medium. The xerographic subsystem can be controlled by a marking engine controller 90, such as a CPU, that includes an actuator for controlling each of the subsystems. The marking engine controller 90 is linked to the image processing system 28 and to the CCS 22 and can also be linked to other system components. In the illustrated embodiment, each of the marking engines 32, 34 has its own CCS 22 that includes a TRC for the respective marking engine, but a single CCS can be used spatially for multiple marking engines. Corrected corrected input values can also be determined and this single CCS can be associated with a common image processing system 28.

Marking engines 32, 34 operate on binary data 30 from image processing system 28. Specifically, the C _in , Y _in , and M _in values 30 are input to the appropriate TRCs 16,92. The input value 30 is selected to achieve the desired target value 94 that will be output in the form of an image, but without taking into account possible colorant interactions. The C _mod , Y _mod , and M _mod values 96 output from the TRC determine the appropriate colorant level for the pixels applied by the image application subcomponents 40, 42 to take into account the colorant interactions. Used when. The image application subcomponent can then create a color document in a single pass or multiple passes.

CITRC 16 can be determined by conventional methods. For example, in the value of M _in and Y _in the selected value of C _mod which corresponds to the set of values of C _in it is to print the calibration patches on the printing medium by the marking engine using separate input value And by measuring the corresponding output value using a suitable sensor such as a colorimeter or spectrophotometer. Regular updates to the CITRC can be made by further measurements or by adjusting the CITRC according to predictions. For example, each of the CITRCs 16 can be stored as one or more basis vectors and appropriate weights that can be modified to accommodate actual or predicted changes.

FIG. 4 illustrates an exemplary method for employing a spatially varying color indexed TRC to determine a modified input value, such as a desired amount of PCS, when used in the presence of an SCS. ing. The order of the steps can be changed from the order shown, and fewer or more steps can be employed. The method starts at step S100. In S110, a spatially varying set of CITRCs 16 is constructed and stored for each of the color separations used in the printing system, for example by being entered into the memory 92 associated with the CCS 22. The In S112, image processing system 28 selects selected cyan (C), magenta (M), and yellow (Y) input values C _in , M for a pixel of interest in a given region 14 of image 10. _in and Y _in are input. The method includes determining a modified input value (C _mod ) for cyan in the presence of a second colorant of magenta and yellow for pixel 12. Similar indexing is performed for M _mod and Y _mod . As an example, the _calculation of C _mod will be described.

In S114, CITRCs four _C in _{-C mod} a maximum which defines the range of values of _{M in} and _{Y in} the desired SCS are identified. These CITRCs are called interpolation nodes. In the case of three color separations, there are two SCSs to consider and thus up to four interpolation nodes. For example, in the embodiment shown in Table 1, if the selected magenta input value (M _in ) is 150 and the selected yellow input value (Y _in ) is 170, then 5; Closed fixed magenta (M) and yellow (Y) input values with CITRCs identified as 6, 9, and 10 above and below the desired magenta (M) and yellow (Y) input values Is selected as an interpolation node. For each of these interpolation nodes, a corresponding cyan modified input value (C _mod ) for the selected cyan input value (C _in ) is obtained (S116). If CITRC is _{sparse, C mod} can be obtained by linear interpolation, weighted average value of values for _{C mod} two _{C in} defining a range of values of the selected _{C in} obtained. In such interpolation method, the weighting applied to each of the values of the two C _mod to be interpolated, the distance value of the selected C _in from the values of the two C _in defining the scope of the CITRC Inversely proportional to Other methods for determining C _in from the CITRC of the interpolation node are also conceivable, such as a method by curve fitting. If there are conserved CITRCs that closely match the desired value of M _in and / or Y _in , fewer than four interpolation nodes can be used.

In S118, the four C _mod values obtained from the four interpolation nodes are interpolated to generate a single combined C _mod value. This interpolation is the distance of the M _in and Y _in values from the fixed M _in and C _in values, where the weight applied to each C _mod value is used in creating the CITRC of the interpolation node. Bilinear interpolation that is inversely proportional to These weightings can be normalized so that the sum is 1.

In S120, when one or more regions exist in the image, for each of the C _mod value and the C _mod value obtained in S118 for the region 14 where the pixel 12 of interest is located, respectively. Interpolation is performed using the value of C _mod similarly obtained for adjacent regions by applying appropriate weightings. These weightings can be normalized. S118 and S120 can be combined using an appropriate function with weights for each of the CITRCs in that region and adjacent regions.

The output of S120 is a single C _mod value that can be used by the image rendering device as the actual input value used for the pixel of interest in the image to achieve the target value C _out (S122). Alternatively, if appropriate, the value of C _mod can be further manipulated. In many common imaging devices, the raster input value is halftone before actually driving the imaging device.

The input values for M _mod and Y _mod for the pixel of interest are determined in a similar manner using a similar set of CITRC for magenta (M) and yellow (Y), respectively (S114-S122). This method can be repeated for each pixel in the image.

  Color indexed TRCs can be represented in the form of basis vectors that can be combined to represent each TRC. In one method, a set of CITRCs as described above is created and a subset of basis vectors representing the CITRC (eg when combined with appropriate weighting) is selected from a larger set of basis vectors. , Dealing with colorant interference. A CITRC that corrects the colorant obtained from the subset of basis vectors is then applied to the digital image to correct for the colorant interference that occurs when the image is rendered by the imaging device.

  Thus, where spatial correction is also performed, a method of employing a spatial and colorant-correcting TRC to deal with colorant interference that varies throughout the space of the desired uniformity is desired uniformity. Creating a set of CITRCs over the entire space, selecting a subset of basis vectors representing those CITRCs, and then applying CITRC to the digital image to correct the uniformity obtained from the subsets of basis vectors Steps. The desired uniform space may be as described above.

  The CITRC can be obtained from a subset of orthogonal basis vectors. The weight can be associated with a basis vector. For example, CITRC for correcting colorants can be applied to digital images as arithmetic operations that form a linear combination of basis vectors.

  In one aspect, the orthogonal basis vector may be a data-dependent basis vector, such as an SVD (Single Value Decomposition) basis vector or a PCA (Principal Component Analysis) basis vector. Alternatively, a base vector that does not depend on data, such as a DCT (Discrete Cosine Transform) basis vector, may be used. A CITRC that corrects the colorant can be applied to the digital image as a look-up table (LUT), where the LUT is created by a linear combination of a subset of basis vectors. Another LUT can be used to identify an appropriate set of CITRCs for a given pixel location in the digital image. The pixel location LUT can use the pixel column and / or row of a given pixel in the digital image as an index into the LUT. The pixel position weighting LUT can be used to select the basis vector weighting for a given pixel in the digital image.

Two or more marking engines may be used to maintain consistency in overall system tone reproduction, such as in printing systems such as those described above where a job page is distributed among similar marking engines. If desired, the CITRC can be adjusted so that the values of C _mod , Y _mod , and M _mod for one marking engine do not exceed the maximum values available to any marking engine in the printing system.

  Optionally, the colorant correction system 22 includes a function that cancels a spatially dependent correlation that corrects aspects based on detected or separated colorant interactions of the measured spatial uniformity variation; Receives a spatially dependent correlation cancellation function from a spatial uniformity detector / isolator based on the colorant interaction, receives a first color and position description, and receives positional information through the correlation cancellation function. A colorant that operates to process the first color description and thereby correct the colorant description to be corrected for the effects of spatially dependent colorant interactions of at least one image rendering device And a spatial uniformity variation compensator based on the interaction.

  The following examples illustrate the effectiveness of the techniques described herein.

  Spatial color correction was performed using multiple color indexed TRCs and compared to values generated by a three-dimensional lookup table. From the three-dimensional lookup table (9 × 9 × 9 lookup table) where each point in the lookup table represents a fixed value of cyan (C), magenta (M), and yellow (Y) For each of cyan (C), magenta (M), and yellow (Y) in the first colorant at a fixed level of the second colorant, the 16 color indexed TRCs are Created. This process was repeated for eight more three-dimensional look-up tables, each corresponding to a separate location of the process on the marking engine. Each created CITRC was composed of 9 input values corresponding to 9 modified input values representing color separation in the presence of a fixed value of the SCS. As an example, a TRC for the zero position (first position) is given in FIG. There is a deviation from the three-dimensional lookup table when a color indexed TRC is applied to the input value of the target CMY colorant to obtain a corrected input value of the color, but this deviation is indexed by CITRC. Can be minimized by optimizing the selection of points and by increasing the number of stored values in each of the TRCs. The advantage of a color indexed TRC over a 3D look-up table is that it saves significant memory. In this example, a 9 × 9 × 9 look-up table requires 9 × 9 × 9 × 3 = 2187 bytes of memory, but 48 CITRCs require only 48 × 9 = 432 bytes of memory. do not do. In short, the amount of memory required is over 80%.

It is a figure which shows the top view which looked at the image from the top, and the exemplary tone reproduction curve regarding one area | region of the image. FIG. 3 illustrates a tone reproduction curve specific to an exemplary colorant according to an aspect of an exemplary embodiment. 1 is a block schematic diagram illustrating one embodiment of a printing system according to an exemplary embodiment. FIG. 4 illustrates example steps of a method for determining an input value of a colorant according to an example embodiment.

Explanation of symbols

  10 images, 12 pixels, 14 regions, 16 color indexed TRC, 20 digital printing system, 22 color correction system, 24 image data, 28 image processing system, 30 binary data, 32, 34 output device, 36 input terminal, 40 , 42 Xerographic subsystem, 90 marking engine controller, 92 memory, 94 target values, 96 values.

Claims (4)

A method for converting colors,
Regarding a first color separation of a plurality of color separations in a digital image,
a) establishing a plurality of tone reproduction curves, wherein each of the plurality of tone reproduction curves is a first colorant for rendering the first color separation; and at least one of the plurality of color separations. Representing the interaction of the colorant with at least one second colorant rendering two second color separations, each of the plurality of tone reproduction curves being a plurality of fixed of the at least one second color separation. Each of the plurality of tone reproduction curves corresponding to each of the inputted input values, wherein the at least one second color separation has the fixed input value corresponding to the tone reproduction curve. Representing a modified input value group corresponding to one color separation input value group ;
b) a predetermined input value representing said first color separation, with respect to said digital image pixels having a predetermined input value that represents at least the second color separation, from said plurality of tone reproduction curves, the tone reproduction a step wherein the fixed input values of the second color separation corresponding to the curve, at least one selection of the tone reproduction curve closer to the predetermined input value representing the second color separation,
For each c) the selected tone reproduction curves, determining the modified input value of the first color separation corresponding to the predetermined input value representing said first color separation,
if d) two or more tone reproduction curve is selected, by more than one function of the modified input value of the selected tone reproduction the first color separation determined for curves, said about the pixel Determining one interpolated modified input value representative of the first color separation. A method of converting color according to claim 1, wherein each of the plurality of tone reproduction curves, the modified input value group corresponding input value group and represented by the tone reproduction curve, singular value decomposition, principal component analysis Or a method for converting a color represented by a plurality of basis vectors obtained by performing discrete cosine transform . A system for converting colors,
A memory for storing a plurality of tone reproduction curves for each of the plurality of color separations, wherein each of the plurality of tone reproduction curves renders a first color separation of the plurality of color separations ; And at least one second colorant that renders at least one other color separation of the plurality of color separations, each of the plurality of color reproduction curves representing the color tone Each of the plurality of tone reproduction curves corresponding to each of a plurality of fixed input values of a second color separation that is at least one other color separation of the first color separation that is the color separation that is the color separation relating to the reproduction curve. For the input value group of the first color separation when the at least one second color separation has the fixed input value corresponding to the tone reproduction curve. It represents the modified input value group to a memory,
Regarding the pixels of the digital image,
The memory is accessed and, among the plurality of fixed input values of the at least one second color separation, fixed close to a predetermined input value representing the at least one second color component value at the pixel. Identifying at least one tone reproduction curve corresponding to the input value ,
For each of the identified tone reproduction curves, indexing the modified input value of the first color separation which corresponds to a predetermined input value representing said first color separation in the pixels,
If more than one tone reproduction curve is identified by a function of the identified tone two or more of said modified input value of the first color separation is indexed with respect to each reproduction curves for the pixel And a processing component for determining one interpolated modified input value representative of the first color separation. A method for imaging color, comprising:
For each of the pixels of interest in the digital image,
The method according to claim 1, wherein one color separation of a plurality of color separations in the digital image is the first color separation, and another color separation of the plurality of color separations in the digital image. Determining one of the corrected input values for the one color separation by performing the second color separation as
The modified input of the other color separation by performing the method of claim 1 as the first color separation and the one color separation as the second color separation. Determining one value;
It said one by using the modified input value and the corrected input value of the other color separation of color separation, and associated with which colorants and the other color separation associated with at least said one color separation Forming an image using a colorant, and imaging a color.

Images